The MPA-JHU DR7 release of spectrum measurements


January 2010: Added SFRs, oxygen abundances etc.

January 2010: Future updates and software will be found at The MPA page will only be updated when absolutely necessary.

This set of pages contain the derived galaxy properties from the MPA-JHU emission line analysis for the SDSS Data Release 7 (DR7). This represent a significant extension in size relative to our previous DR4 release and there are a number of improvements in the data. This is both due to improvements in the reduction pipeline by the SDSS and to improvments in our analysis pipeline, not least being the adoption of new Charlot & Bruzual (2008) stellar population synthesis spectra for the continuum subtraction.

The data we provide can be divided into two different categories, namely the raw and derived data. The data follow the same layout as described in the previous data release.

The data are described in detail on separate pages - one for the raw data and one for the derived data. This page gives a brief overview of what is where and what is included in this list. Cautions about the data are generally given on the data pages. A total of 927552 galaxy spectra are included in our list, 818333 of these are unique using the definition given below.

Differences from past releases

There are a couple of differences with past releases. These are detailed further below when relevant, but here is a brief outline:

Which plates are included?

The full list of PLATEID and MJD and the number of galaxy spectra on each plate can be found here. And the Yanny format file used for the reduction can be found here

Which objects are included?

On each plate we include all objects that satisfy the following:

The organisation of the data

Raw data

The files with raw data define the order of the objects and are FITS tables that are sorted with the same number of elements. These files contain the line fluxes, equivalent widths and continuum indices as well as information about the objects such as their redshifts, velocity dispersions and plate, fiber and MJD. While these files can be read in directly using mrdfits in IDL, this is not recommended as some of the files are very big. Use other approaches instead.

Go to the raw data

Derived data

In addition to these large FITS tables with raw measurements from the spectra, we also have files with derived data - this include gas-phase metallicity estimates, star formation rates, stellar masses and the These files all correspond one-to-one with the raw files.

At the present time not all these files are available but they will be provided as soon they are ready. Send a note to jarle at if you want to be notified upon release.


The sample included here has a lot of duplicates - this is very useful for assessing the true uncertainties in measured quantities, see in particular the discussion about uncertainty estimates here.

For your convenience, this file contains most of the information necessary for identifying duplicates.

The first column is the index number of each object (in the other files). The remaining columns on each row is a list of all duplicate (close) matches for that object (values <0 indicate no match). There is no cut on velocity. This must be applied by the user - a good rule is to require that the velocity difference (redshift difference) is no greater that 100 km/s or so if one wants a clean sample of duplicate observations